21, rue d’Artois, F-75008 PARIS
http : //www.cigre.org
A2-304
CIGRE 2012
Effects of Geomagnetically Induced Currents on Power Transformers and
Power Systems
R. GIRGIS*, K. VEDANTE
ABB Power Transformers
USA
K. GRAMM
ABB Power Transformers
SWEDEN
SUMMARY
The purpose of this paper is to present to the power industry the true effect of GIC on power
transformers. First, the phenomenon of part – cycle, semi – saturation of transformer cores due to GIC
is described briefly. The magnitude and wave – shape of the resulting magnetizing current pulse are
given for of an example large power transformer when subjected to different levels of GIC. The
corresponding harmonics associated with these magnetizing current pulses are also presented. Proper
evaluation of this magnetizing current pulse and the associated high order harmonics are critical to the
determination of the response of power systems to a GIC event.
In the discussion of the effects of GIC on the temperature rises in windings and structural parts of a
power transformer, typical signatures of GIC associated with Solar Magnetic events are presented.
Based on these measured signatures, calculations are presented in the paper for winding and structural
parts hot spot temperature rises due to a representative GIC signature.
One of the most significant contributions of this paper is a detailed description of some of the cases in
which power transformers were reported to have suffered damage, or over – heating, during past GIC
events which was blamed solely on GIC. Also presented in this paper, are actual measurements, made
by 3 major utilities on full size power transformers when subjected to high levels of DC. The paper
demonstrates that, because of the nature of the GIC currents, the great majority of power transformers
would not experience damaging overheating due to even high levels of GIC. Only transformers with
certain design features could suffer some winding damage due to high winding circulating currents
when exposed to high levels of GIC.
The paper also presents the real issue with GIC; namely, the narrow pulse of the magnetizing current
which results from part – cycle, core semi – saturation when subjected to high levels of GIC currents.
This one current pulse / cycle causes relays and capacitive components in power systems, such as
SVC(s), to trip causing grid instability. Also, the current pulse is associated with higher order
harmonics. As a result, resonance may occur, differential relays may operate, and stability of the grid
may be compromised.
KEYWORDS
Power Transformers, Power Systems, Geo – Magnetically Induced currents, GIC, DC
*ramsis.girgis@us.abb.com
1
1. INTRODUCTION
Effects of Solar Geomagnetic Disturbances on power transformers and power systems have been
observed for many decades and in many parts of the world. These events had created a certain level of
concern in the electric power industry about the implications of GIC currents flowing into the neutrals
of large power transformers. Now, with the approach of the upcoming solar cycle (Cycle 24) in the
Years 2012 – 2013, the level of concern is as great as ever. Additionally, there is some misconception
in the electric power industry today that Geo – magnetically induced currents have caused, and would
cause significant damage to many of the large power transformers installed around the World,
resulting in transformer failures of a large scale.
2. PHENOMENON OF PART – CYCLE, CORE SEMI – SATURATION UNDER
EFFECT OF DC / GIC
When a power transformer is subjected to DC, it results in a unidirectional DC flux in the core. The
magnitude of this flux depends mainly on the magnitude of the DC current, number of turns in the
windings carrying the DC current, and reluctance of the path for this DC flux. The result is that the DC
flux adds to the AC flux in one half – cycle and subtracts from the AC flux in the other half – cycle.
When large enough, this leads to core peak flux densities in the magnetic core pre – saturation range in
one half of the cycle. For higher magnitudes of DC, the core provides a very high reluctance to the DC
Ampere – Turns; resulting in a smaller further increase in the flux density DC shift. Correspondingly,
the magnetizing current becomes a short duration pulse with a high peak. The duration of this pulse is
only in the range of 1/6th to 1/10th of the cycle, Figure 1.
B
AC + DC
AC
Time
I
AC + DC
AC
Time
Fig. 1: Part – Cycle, Semi – Saturation of Transformer cores when subjected to DC / GIC
As mentioned earlier, the magnitude of the DC flux shift in the core depends on the magnetic
reluctance of the DC flux path. It follows that the DC flux shift in a 3 – phase, 3 – limb core form
transformers would be the lowest of all core types. This is because this core design offers orders of
magnitude higher magnetic reluctance to the DC Ampere – turns in the Core – Tank magnetic circuit.
In this core type, the DC flux has to pass through the very high reluctance path from the core top yoke
to the tank cover, through the tank walls, and return to the bottom yoke through again the high
reluctance path from the tank bottom. All other core types offer much less reluctance to the DC
Ampere – turns because the return path to this DC flux is through the core which has orders of
magnitudes higher permeability. Core material and core joint type have some influence. However this
influence is small and depends on the core type and the operating flux density in the core.
2
3. MAGNITUDES AND HARMONIC CONTENT OF THE MAGNETIZING
CURRENT PULSE
As explained above, magnitudes of the magnetizing current pulse resulting from core semi – saturation
are typically an order of magnitude higher for transformers with core types other than the 3 – phase, 3
– limb core form. It follows that magnitudes of the harmonic content are also lowest for 3 – phase, 3 –
limb cores. This is demonstrated in Figure 2 below. The Figure presents the % harmonic content of the
magnetizing current of two transformers of two different core – types, relative to their corresponding
full load current, when subjected to 30 Amps DC / phase. The figure shows that the short duration
nature of the magnetizing current pulse results in an almost uniform magnitude of low and high order
harmonics. Another observation from Figure 2 is that transformers have levels of the 2nd order
harmonic that would give an erroneous message to the differential relay set at a low value of 2nd
harmonic content to differentiate between Inrush current events and a fault condition. Additionally, as
will be explained in section 8 of this paper, the 3rd order harmonics could cause relays to operate
resulting in tripping of transmission lines during a GIC event; hence grid instability and system
blackouts.
Hamonic Spectrum of Magnetizing Current with same DC Current/Phase
6%
1-Phase, 3 Limb
Harmonic Amplitude, % of Rated Current
3-Phase, 3 Limb
5%
4%
3%
2%
1%
0%
60
120
180
240
300
360
420
480
540
600
660
Harmonic Frequency, Hertz
Fig. 2 Harmonic content of magnetizing current of 2 transformers subjected to DC / GIC
4. INCREASES IN HOT SPOT TEMPERATURES OF WINDING AND
STRUCTURAL PARTS WHEN SUBJECTED TO DC / GIC
The several orders of magnitude higher magnetizing current, and the nature of its wave – shape,
produce correspondingly higher magnitudes of leakage flux that is also rich in harmonics. This results
in appreciably higher eddy and circulating current losses in the windings and the structural parts of the
transformer; increasing their temperatures, Figures 3 and 4. The figures show that the hot spot
temperature reached close to a final value within 5 minutes for windings, and within 10 minutes for
structural parts, from the application of the DC. This corresponds to the typical range of the time –
constant of the windings and structural parts, respectively. Under actual GIC, however, this
temperature rise will be much lower than shown below due to the very short duration nature of GIC (1
– 3 minutes). This is demonstrated in the following section of this paper.
3
Winding Hot Spot Temperature vs Time, 1-Phase Transformer
130
120
110
Wdg Hot Spot Tempt, Degree C
100
90
80
Idc = 50 Amps
70
Idc = 30 Amps
60
Idc = 20 Amps
50
40
30
20
10
0
0
5
10
15
Time, Minutes
20
25
30
100
120
Fig. 3 – Winding Hot spot temperatures for different values of DC.
Flitch-Plate Temperature vs Time
160
140
Flitch-Plate Tempt, Degree C
120
100
Idc = 20 Amps
Idc = 30 Amps
80
Idc = 50 Amps
60
40
20
0
0
20
40
60
Time, Minutes
80
Fig. 4: Tie – Plate hot spot temperature for different values of DC Current
4
5. EFFECT OF GIC ON POWER TRANSFORMERS
In order to accurately determine the capability of a transformer to GIC, one needs to consider the
nature of the signature of the GIC current; namely its magnitudes and duration. GIC current is
typically characterized by low levels of 10’s of minutes / a few hours duration interrupted by a few,
much higher, peaks of pulses that have a very short time duration (typically 1 – 3 minutes).
Due to the short duration of these high peaks of GIC and the fact that the duration of the resulting
magnetizing current pulse is only a small fraction of a cycle, the actual duration of the resulting core
part – cycle, semi saturation and associated high peak pulses of the core magnetizing current is only a
few seconds. Hence, temperature rises in the transformer windings and structural parts due to GIC
would be expected to be much lower than that estimated based on continuous duration DC currents. In
order to illustrate the above, a profile of GIC, per Figure 5 (a) below, was assumed and winding hot
spot temperature was calculated for a fully loaded large 1 – phase power transformer for this assumed
GIC current profile. The assumed GIC profile has a base level of 20 Amps / phase (60 Amps for a 3
phase bank) interrupted, every 30 minutes, by a 2 – minute duration very high level GIC pulse of 400
Amps / phase (1200 Amps for a 3 phase bank). The results are presented in Figure 5 (b).
It can be seen from the figure that:
1. The temperature rise of the winding hot – spot due to the base 20 amps is only 1º C.
2. The temperature rise of the winding hot – spot, due to the 2 – minute duration of 400 Amps /
phase of GIC, increases by about 35 C; and a hot spot total temperature of about 146 C. The
temperatures would be much lower when the transformer is not fully loaded and / or ambient
temperature is lower than 30 C. Such winding hot spot temperatures for such a short duration
would not cause any appreciable damage of the windings, structural parts, or loss of insulation
life of the transformer. In fact. Industry Standards for overload (IEEE and IEC) allow much
higher winding hot spot temperature levels for much longer times under emergency loading
conditions (160 – 180 C).
3. After these 2 minutes, the hot spot temperature of the windings goes back down to the original
temperature existing before that pulse of GIC.
4. The duration of that pulse of temperature rise of the windings is only a few minutes; which
would not result in any damage to the windings.
The same is true for structural parts, except that the temperature rise will be even smaller for these
structural parts as their time constant is higher than that of the windings.
400 Amps
400 Amps
Idc
20 Amps
20 Amps
20 Amps
(0,0)
30
2
30
2
30
Time, Minutes
(a) Simulated GIC Current
5
Winding Hot Spot Temperature vs Time
160
Winding Hot Spot Tempt, Degree C
140
120
100
80
60
40
20
0
0
10
20
30
40
50
60
70
80
Time, Minutes
(b) – Calculated Winding Hot Spot Temperature
Fig. 5 – Winding Hot Spot Temperature for Simulated GIC profile per Fig 10 (a)
6. TRANSFORMERS REPORTED AS HAVING FAILED / SUFFERED
OVERHEATING DURING GIC EVENTS
Reference [1] reports several of those cases. One of these is a shell form transformer that experienced
significant overheating of the LV leads of a GSU transformer at PSE&G in NJ, USA, during the
March 13, 1989 K9 GIC event. The transformer was taken out of service a week later because of
significant gassing. Same design transformers at the same general location experienced similar
overheating, although to a lesser degree. These were old Shell form transformers that had an old
winding lead design that made it susceptible to overheating caused by high circulating current when
the core experienced part – cycle, semi – saturation; during which the leakage flux pattern changed
and resulted in very high circulating currents in the LV windings of this transformer [2].
Another case reported in Reference [1] is a Shell – form Transformer that had tank wall heating during
the same GIC event. In such transformers, wood slabs are placed between the core and tank walls.
When the core goes through semi – saturation, part of the core and leakage flux travel to the tank wall
causing localized eddy losses in the tank. Being blanketed by the wood slabs, these regions of the tank
overheated. Additionally, it should be noted that this temperature increase had only resulted in
discoloration of the tank paint at the tank regions opposite to the wood slabs. New designs allow
improved cooling of tank walls in these regions.
It was also reported in Reference [1] that within 2 years of the 1989 GIC event, 11 nuclear plants
experienced failures of several GSUs. During that period, a number of these failures were studied and
found to be caused by back – feed mode operation. In this mode of operation, the generator step – up
transformers are not sufficiently protected from switching and lightning surges and the electrical
damping in the electric circuit is very low making it vulnerable to transient voltage magnification due
to winding resonances.
Other cases were reported in Reference [3] of significant winding overheating in a few large core form
power transformers in S. Africa during the period between 2003 and 2004. These incidents were found
to coincide with winding overheating caused by the phenomenon of the conducting Copper Sulphide
6
forming and causing winding failures of transformers worldwide. This was attributed to high sulphur
content in some types of transformer oil during this period. It is plausible that increases in the winding
temperature, caused by GIC, could have accelerated the formation of Copper Sulphide in the windings
of these transformers. However, the magnitudes of GIC, reported to have been associated with these
incidents, do not seem to be high enough to cause much winding overheating.
It was reported by HQ in reference [4] that two transformers were subjected to external faults during
the GIC event and are believed to have been caused by over – voltages due to System instability Also,
one transformer in New Zealand, that was reported in the same reference as have failed one minute
after a GIC event, is believed to have been caused by System instability associated with the GIC event.
Also, Reference [5] reports on two transformer failures in National Grid in Britain after a GIC event
but no details were given to ascertain that claim. Finally, in Reference [6], account is given of a
number of 400 kV large power transformers in southern Sweden that were subjected to high levels of
GIC in the October 2003 GIC event. No transformer damage was reported; including a 3 – Phase, core
– form transformer with a 5 – limb core, which was subjected to an estimated peak of 330 Amps of
GIC in the neutral. The authors of this paper reported that a 20 – 50 minute black – out occurred in the
Malmo area which was caused by the tripping of a 130 kV line resulting from the operation of a relay
that had a higher sensitivity to the 3rd harmonic than to the fundamental frequency.
It is to be noted here that, as in the case of the PSE&G shell form transformers, core and shell form
transformers, with certain winding / lead arrangements, could experience high levels of circulating
current and suffer some winding overheating when these transformers are subjected to levels of GIC
that would cause core semi – saturation. These types of transformers will need to be investigated and
identified on an individual basis.
7. ACTUAL MEASUREMENTS OF EFFECT OF DC / GIC
Two full size experiments were performed by Hydro Quebec [7] and FIN Power [8], where very large 1
– phase and 3 – phase power transformers were injected with high levels of DC (75 Amps for 20
minutes and up to 200 Amps for 30 minute intervals), respectively. Temperature rises were measured
in windings and structural parts of the transformers. It was found that the temperature rises were in a
reasonable range in spite of the high levels of DC applied continuously for long durations. As
concluded by these studies, even high levels of GIC currents should not cause transformer damage.
An earlier experimental study of the effect of DC on power transformers was conducted by Tokyo
Electric Power Corporation, in collaboration with Toshiba, Hitachi, and Mitsubishi [9]. The study
tested several small scale models of core form as well as shell form transformers with different core
types. Based on the measurements performed on these models, measurements were made on two large
– Scale models of 1 – phase, 3 – legged 1000 MVA / 550 KV core form and shell form transformers.
These transformers were tested with up to a DC level of 66 Amps which corresponds to 400 – 600
Amps / phase for the corresponding full size transformers. The DC was applied for up to 20 – 30
minutes continuous. The leakage flux and temperatures were measured in windings and structural
parts of these transformers. The maximum temperature rise measured after 30 minutes of applying
such large magnitudes of DC continuously was approximately 110 C in both the Tie – plate for the
core form transformer and the core support in the shell form transformer. The study again concluded
that because of the short duration of these temperatures, the transformer life would not be appreciably
affected. Also, the very short duration of GIC would not allow hot spot temperatures to rise to a
fraction of these temperatures.
8. EFFECT OF GIC ON POWER SYSTEMS
As presented earlier in this paper, part – cycle, core semi – saturation is associated with a high
magnitude narrow magnetizing current pulse; once per cycle. This represents high levels of VAR
pulse requirement on the power system causing system voltage distortions. Also, the narrow current
pulse causes capacitive components on the system, such as static VAR compensators, etc. to increase
their currents and may become overloaded and trip causing grid instability. Moreover, the magnetizing
current pulse is associated with high harmonics, and as a result, resonance may occur and stability of
the grid may be compromised due to the creation of virtual zero at some point and opening of
7
transmission lines. Also, as observed from Figure 2 of this paper, transformers have levels of the 2nd
order harmonic. This is significant as lower levels of 2nd order harmonics associated with the total
primary current of the transformer would give an erroneous message to the differential relay set at a
low value of 2nd harmonic content to differentiate between Inrush current events and a fault condition.
Additionally, as reported in Reference [6], operation of relays that have higher sensitivity to the 3rd
harmonic than to the fundamental frequency, would cause tripping of transmission lines during a GIC
event; causing grid instability and system blackouts.
9. CONCLUSIONS
 High levels of GIC could cause part – cycle, semi – saturation of transformer cores. This results in
a short duration, high peak pulse of magnetizing current.
 Because of its short duration, in the great majority of the cases, even high levels of GIC would not
cause damaging overheating of neither windings nor structural parts of power transformers.
 Only transformers, core – or shell – form, where part – cycle, core semi – saturation would cause
high winding circulating currents would be susceptible to winding overheating when the
transformer is subjected to high levels of GIC.
 Cases of significant overheating and winding damage, reported in the published literature as have
been solely caused by GIC, were found to have been caused totally, or partially, by other effects,
or by system instability experienced during or after the GIC event.
 The main impact of GIC is on the system instability it causes due to high levels of VARS and
significant current harmonics as a result of transformer part – cycle, core semi – saturation.
10. BIBLIOGRAPHY
[1] John Kappenman, “Geomagnetic Storms and Their Impacts on the U.S. Power Grid”, Report #
Meta-R-319 by Metatech Corporation submitted to Oak Ridge National Laboratory, Pages 1-197.
[2] Ramsis Girgis & Chung-Duck KO, “Calculation Techniques And Results Of Effects Of GIC
Currents as Applied to Two Large Power Transformers”, IEEE Transactions on Power Delivery, Vol.
7, No. 2, April 1992, Page 699-705.
[3] C. T Gaunt, G. Coetzee, “Transformer failures in regions incorrectly considered to have low GICrisk”, Mat Post 07, 3rd European Conference on MV & HV Substation Equipment, Nov 15-17, 2007,
Lyon, France, Proceedings of Power Tech, July 15, 2007, Lausanne, Switzerland
[4] Jean Beland, Kevin Small, “Space Weather Effects on Power Transmission Systems: The Cases of
Hydro-Quebec and Trans power New Zealand Ltd,” Proceedings of the NATO Advanced Research
Workshop on Effects of Space Weather on Technology Infrastructure, NATO Science Series, Vol.
176, 2004
[5] A. Erinmez, J. G. Kappenman, W. A. Radasky: “Management of the geomagnetically induced
current risks on the National Grid Company’s electric power transmission system,” Journal of
Atmospheric and Solar-Terrestrial Physics, 64, 2002, p743-756
[6] M. Wik et al, “Space weather events in July 1982 and October 2003, and the effects of
geomagnetically induced currents on Swedish technical systems”, Annales Geophysicae, 27, 17751787, 2009.
[7] P. Picher, L. Bolduc et al, “Study of the acceptable DC current limit in core-form Power
Transformers”, IEEE Transactions on Power Delivery, Vol. 12, No. 1, January 1997, Page 257-265.
[8] Matti Lahtinen & Jarmo Elovaara, “GIC Occurrences and GIC Test for 400 kV System
Transformer”, IEEE Transactions on Power Delivery, Vol. 17, No. 2, April 2002, Page 555-561.
[9] Nobuo Takasu et al, “An experimental analysis of DC excitation of Transformers by Geo magnetically Induced Currents”, IEEE Transactions on Power Delivery, Vol. 9, No. 2, April 1994,
Page 1173-1182.
8